Hi all,
I am wondering if there is any discussion or plan to implement distribution class in MxNet, similar to tensorflow.contrib.distributions (see also their recent paper).
There are a few teams also in progress to implement something similar in PyTorch. You can find their discussion of the design here: https://goo.gl/9ccYsq.
Hello, a colleague of mine, asmus hetzel, implemented several of the sampling ops in my.nd.random.
It would be very useful to beef this up and organize it more along distributions. Unfortunately, we do not find the time to do it.
1 Like